Movies Project

library(tidyverse)
library(moderndive)
library(infer)
metadata = read_csv("clean_metadata.csv")

Missing column names filled in: 'X1' [1]Parsed with column specification:
cols(
  X1 = col_double(),
  id = col_double(),
  imdbId = col_character(),
  movieId = col_double(),
  tmdbId = col_double(),
  original_title = col_character(),
  title = col_character(),
  popularity = col_double(),
  release_date = col_date(format = ""),
  revenue = col_double(),
  runtime = col_double(),
  budget = col_double(),
  vote_average = col_double(),
  vote_count = col_double(),
  original_language = col_character()
)

metadata  = metadata %>% select(-X1)
metadata

metadata %>%
  arrange(desc(budget))

metadata = metadata %>%
  separate(col = release_date, into = c("year", "month", "day"), sep = '-')
metadata

metadata %>% 
  group_by(month) %>%
  summarize(median_revenue = median(revenue, na.rm = T)) %>%
  arrange(desc(median_revenue))

NA

metadata %>% filter(revenue < 1.5e+9) %>%
  ggplot(aes(month, revenue, fill = month)) + geom_boxplot(show.legend = F) + coord_flip()

metadata %>%
  ggplot(aes(budget, vote_average)) + geom_point() + geom_smooth() + geom_smooth(method = 'lm', color = 'tomato') + ggtitle("Average Vote vs. Budget") + xlab("Budget") + ylab("Average Vote")

metadata %>%
  arrange(desc(revenue)) %>%
  select(title, revenue, budget, year) %>%
  head()

metadata %>%
  filter(budget > 2e+8) %>%
  arrange((revenue)) %>%
  select(title, revenue, budget, year) %>%
  head()

metadata %>% 
    get_correlation(revenue ~ budget, na.rm = T)

metadata %>% filter(popularity < 150) %>%
  ggplot(aes(popularity, vote_average)) + geom_point() + geom_smooth()

metadata %>%
  ggplot(aes(vote_average, revenue)) + geom_point() + geom_smooth() + geom_smooth(method = 'lm', color = 'tomato')

metadata %>%
  ggplot(aes(budget, vote_average)) + geom_point() + geom_smooth() + geom_smooth(method = 'lm', color = 'tomato')

metadata %>% filter(popularity < 50) %>%
  ggplot(aes(popularity, vote_average)) + geom_point() + geom_smooth() + geom_smooth(method = 'lm', color = 'tomato')

metadata %>% 
  arrange(desc(popularity))

metadata %>%
  ggplot(aes(popularity, vote_average)) + geom_point() + geom_smooth() + geom_smooth(method = 'lm', color = 'tomato')

metadata %>% filter(original_language == "ja" | original_language == "it" | original_language == "fr" |original_language == "en" ) %>% filter(revenue < 1e+9) %>% 
  ggplot(aes(original_language, revenue, fill = original_language)) + geom_boxplot() + coord_flip()

metadata %>%
  arrange((runtime))

metadata %>% filter(revenue < 1e+9) %>%
  ggplot(aes(month, revenue, fill = month)) + geom_boxplot(show.legend = F) + coord_flip()

metadata %>% filter(budget < 2e+8) %>%
  ggplot(aes(budget, vote_average)) + geom_point() + geom_smooth()

budget_vote_average_mod = lm(vote_average ~ budget, data = metadata)
budget_vote_average_mod %>%
  get_regression_table()

yhat = 6.46 + 0x

metadata %>% 
    get_correlation(vote_average ~ budget, na.rm = T)

-0.08975617 * -0.08975617

[1] 0.00805617

metadata %>%
  ggplot(aes(month, vote_average, fill = month)) + geom_boxplot(show.legend = F) + coord_flip()

metadata %>%
  ggplot(aes(popularity, popularity)) + geom_point(show.legend = F)

metadata %>%
  ggplot(aes(budget, revenue)) + geom_point(show.legend = F) + geom_smooth() + geom_smooth(method = "lm", color = "tomato")

metadata %>%
  arrange(desc(popularity))

metadata %>% 
    get_correlation(revenue ~ budget, na.rm = T)

0.7091052 * 0.7091052

[1] 0.5028302

budget_revenue_mod = lm(revenue ~ budget, data = metadata)
budget_revenue_mod %>%
  get_regression_table()

yhat = 2998573.384 + 3.007x